* * * CHINESE OCR API * * *
Chinese OCR API.
Contracts, receipts, scanned books, screenshots of WeChat — modern vision-language models read Chinese the way they read English, no language pack or config flag required. POST the file, get 中文 back as text or markdown, Simplified and Traditional alike, even when the page mixes in English or numbers.
START FREE — 100 PAGES
$0.75 / 1,000 PAGES
1,333 PAGES PER DOLLAR · NO CREDIT CARD
01 / TRY IT
ONE ENDPOINT.
POST a file, get JSON back — the extracted text, per page and joined. PDF, PNG, JPEG, WebP or TIFF.
curl https://api.pennyocr.com/v1/ocr \
-H "Authorization: Bearer $PENNYOCR_API_KEY" \
-F "file=@contract-zh.pdf"
# $0.75 per 1,000 pages, first 100 free02 / USE CASES
WHAT PEOPLE SEND.
INVOICES & FAPIAO
Read supplier invoices and fapiao into text your accounting pipeline can parse. Line items, amounts and tax fields in reading order.
CONTRACTS & FILINGS
Digitize scanned Chinese contracts and registry documents — dense pages, chops and stamps included.
MIXED-LANGUAGE DOCS
Bilingual contracts, export paperwork, packing lists — Chinese and English on the same page come back in one clean pass.
BOOKS & ARCHIVES
Batch-scan vertical or horizontal layouts into searchable text at $0.75 per 1,000 pages.
03 / PRICE CHECK
HALF THE PRICE OF THE BIG CLOUDS.
Per 1,000 pages, public list prices, first tier.
PENNYOCR$0.75
AWS TEXTRACT$1.50
GOOGLE CLOUD VISION$1.50
AZURE DOC INTELLIGENCE$1.50
YOU KEEP50%
SIMPLIFIED OR TRADITIONAL?
Both. The model reads whichever script is on the page and returns it as-is — it never silently converts one to the other.
DO I NEED TO TELL YOU THE LANGUAGE?
No. There is no language parameter — the model detects Chinese, English, or a mix on its own, per page.
HOW ACCURATE IS IT ON CJK TEXT?
Vision-language models handle CJK far better than classic OCR engines, which lean on Latin-centric heuristics. Real users run Chinese documents through PennyOCR daily — send your hardest scan on the free tier and check the output yourself.
ALSO ON THE MENU